Does one's own and one's spouse's education affect overall and cause-specific mortality in the elderly?

OBJECTIVES: To examine educational gradients in overall and cause-specific mortality among elderly married men and women and their spouses. METHODS: Using the census-based Israel Longitudinal Mortality Study (1983-92), 13 573 married men and 6563 married women were identified who were aged 70-89 years at baseline. Cox proportional hazard models were used to assess the strength of the association between education and overall and cause-specific mortality. RESULTS: Educational gradients for own and spouse's mortality varied by gender and cause of death. In particular, in relation to cardiovascular disease, men married to uneducated wives experienced elevated mortality risks [hazard ratio (HR) = 1.30; 95% confidence interval (95% CI) 1.11-1.52]. Women were generally unaffected by their husband's education, except for those who died from non-breast cancer, for whom husband's low education had a harmful effect (HR = 1.98; 95% CI 1.26-3.11). CONCLUSIONS: Mortality among elderly married persons is associated with one's own and one's spouse's educational achievement. Research using partner's education as a proxy for one's own attainment may be omitting valuable information regarding these and other health risks.

Alcohol drinking patterns among Jewish and Arab men and women in Israel.

OBJECTIVE: Jews and Muslim Arabs comprise the bulk of modern Israeli society. Jewish tradition permits controlled alcohol drinking, whereas Muslim tradition prohibits the use of any alcohol. Increasing exposure of the traditionally conservative Arab sector to the Western culture of modern Israel might impact on and be reflected in the drinking patterns of these two populations. The influence of religiosity and other factors on drinking patterns of Jewish and Arab adults are examined using data from a 1995 national household survey. METHOD: Past month drinking is assessed in this nationally representative sample of nearly 5,000 Jews and 1,000 Arabs (N = 5,954, 60% women). Unadjusted and adjusted odds ratios (ORs) are presented to describe associations between any and heavy drinking and nationality group, religiosity, education and marital status among men and women. Modification of the nationality-drinking relationship by religiosity is also examined. RESULTS: Any past-month drinking was reported more often by Jewish respondents than Arab respondents (OR = 2.9, 95% Cl: 2.5-3.4), and this difference remained statistically significant after accounting for the effects of the other covariables. This cross-nationality difference was more pronounced among women (OR = 6.4, 95% Cl: 4.6-8.8) than men (OR = 2.3, 95% CI: 1.8-2.9). The proportion of drinkers who reported heavy drinking in the past month, however, was lower among Jews (OR = 0.3, 95% CI: 0.2-0.4). Significantly higher rates of drinking were noted for secular men and women than for religious respondents in both nationality groups. Rates of drinking were more similar among secular Arabs and Jews than among religious respondents of these nationality groups. CONCLUSIONS: These results add support to the theory that adherence to religious traditions continues to serve as a barrier against drinking among both Arabs and Jews. Further work is required to determine if these patterns are stable over time and whether genetic factors are contributing to the sociocultural influences.

Birthweight recall by mothers of Israeli children.

The objective of this study was to examine the reliability and validity of maternally recalled birthweight (BW) in a population of 6-y-old children in Jerusalem. The study population consisted of 259 children entering the first grade in two schools in the city. We compare the recalled BW at entrance to school and the BW values as recorded in the Mother and Child Health Clinic of the Hadassah Community Health Center. The mean recalled BW was 11.7 g lower than the mean recorded BW (P>0.05). There were no statistical differences in mean BW values between the two data sources (according to socio-demographic characteristics) except for mother's age and origin, where no trends could be detected. Fifty-eight percent of the mothers accurately recalled the BW of their children to within 100 g and 80% recalled the BW to within 500 g. Seventy-three percent of mothers of low-birthweight children recalled a BW value of less than 2500 g and 99% of mothers of normal BW children recalled a BW of 2500 g or more. Correlation coefficients between the two sources ranged from 0.89 to 0.96. Given the high degree of accuracy, maternally recalled birthweight is a reliable and valid measure of birthweight 6 y post-partum and its use is justified when no records of birthweight are available. Public Health (2000) 114, 161-163

Drug use among welfare recipients in the United States.

This study examined the prevalence of drug use in a nationally representative sample of 1989 recipients and 6840 nonrecipients of four welfare programs. Data from the 1995 National Household Survey on Drug Abuse (NHSDA) were analyzed using the conditional form of multiple logistic regression with matching of respondents on neighborhood of residence. Weighted proportions and variances accounting for the complex sample design of the NHSDA survey were estimated using the Taylor series linearization method. The results indicate that drug use is 50% more common in households with welfare recipients than in nonwelfare households. Programs making welfare eligibility contingent on the recipient working toward a drug-free lifestyle are worth examining, although a vigilant eye must be kept on the potential unintended consequences.

"Drug dependence" and death: survival analysis of the Baltimore ECA sample from 1981 to 1995.

AIMS: Illicit drug use and dependence often are associated with premature death, but available evidence comes mainly from clinical samples. The present paper examines drug-related mortality experience over 14 years in a United States community sample. PARTICIPANTS: Following probability sampling, 3,481 adult community household residents were recruited for the 1981 NIMH Baltimore Epidemiologic Catchment Area survey. Follow-up occurred in 1993-1996. METHODS: Survival analyses were used to estimate median age at death and relative risk of dying in relation to drug use and dependence as assessed in 1981 using the Diagnostic Interview Schedule (DIS). FINDINGS: Cases with DIS "drug dependence" were more likely to have died and to have a younger median age at death (p < .05), with and without statistical adjustment for confounding variables. Higher levels of drug involvement also were associated with increased age-adjusted mortality. CONCLUSIONS: The evidence favors the hypothesis that DIS-elicited "drug dependence," as well as subthreshold drug use, help to account for premature death in this community sample.

"Alcohol dependence" and death: survival analysis of the Baltimore ECA sample from 1981 to 1995.

OBJECTIVE: Evidence is provided about the association between "alcohol-use disorders" and the 14-year risk of death in a community sample. Most prior descriptions of this association come from treatment samples. METHOD: 3,481 adult household residents were recruited into the NIMB Baltimore Epidemiologic Catchment Area survey and interviewed in 1981. The Diagnostic Interview Schedule (DIS) was employed to assess alcohol drinking and other drug-taking behaviors, and to determine fulfillment of DSM-III criteria for "alcohol abuse" and/or "dependence" diagnoses. Participants were followed-up in 1993-1996, by which time 24% of the sample had died. Median age of death was estimated for persons with and without alcohol disorders, and for "heavy" and "nonheavy" drinkers. Cox proportional hazards models adjusted for the influence of age, sex, race, "drug-use disorders," and tobacco smoking. RESULTS: "Alcohol abuse" and/or "dependence" was associated with a higher risk of death and a younger median age of death (adjusted relative risk = 1.3, p = .016). "Heavy" alcohol consumption was also associated with a significantly elevated risk of death. The DIS diagnosis of "alcohol use disorder" helped predict mortality over and above a prediction based solely upon "heavy drinking" (p < .01). CONCLUSIONS: These findings indicate that the observed increased risk of death associated with "alcohol dependence" is not limited to cases severe enough to have been treated but is also present among cases in the household population.

Male-female differences in the earliest stages of drug involvement.

AIMS: To examine whether male-female differences in rates of drug use could be traced back to differences in rates of exposure to initial opportunities to try drugs, rather than to sex differences in the probability of making a transition to use, once opportunity has occurred. DESIGN: Cross-sectional sample survey research with nationally representative samples and retrospective assessments. SETTING: United States, 1979-94. PARTICIPANTS: Respondents were 131,226 residents aged 12 years and older, recruited for the National Household Surveys on Drug Abuse by multi-stage probability sampling at nine time points. MEASUREMENTS: Estimated proportion of males and females with an opportunity to use marijuana, cocaine, hallucinogens and heroin; proportions reporting use among those having an opportunity to use each drug; proportion making a "rapid transition" from initial opportunity to initial use. FINDINGS: For each survey year, males were more likely than females to have an initial opportunity to use drugs. However, few male-female differences were observed in the probability of making a transition into drug use, once an opportunity had occurred. CONCLUSIONS: Previously documented male excess in rates of drug use may be due to greater male exposure to opportunities to try drugs, rather than to greater chance of progressing from initial opportunity to actual use. This suggests that sex differences in drug involvement emerge early in the process. Implications of these findings are discussed in relation to the epidemiology and prevention of drug use, and future research on sex differences in drug involvement.

Familial resemblance of alcohol consumption levels in Jewish families.

The role of genetic and environmental factors determining the variability in alcohol consumption levels was investigated in 68 families ascertained through heroin-dependent Jewish male probands. Sibling correlations for peak weekly alcohol consumption ranged from 0.22 to 0.32, with limited changes on adjustment for sex, age and environmental variables. The parent-child correlations were relatively low. Segregation analysis indicated that a major effect of a non-transmitted environmental factor explained the mixture of distributions. There was no evidence for a polygenic effect on alcohol consumption in the families. When segregation models were fitted to sex, age and environment-adjusted alcohol levels, the mixed environment model was rejected, whereas the mixed genetic model was not. These findings are consistent with two previously published segregation analyses of alcohol dependence, and further highlight the heterogeneous aetiology and transmission of alcohol consumption and alcohol dependence.

The epidemiology of adolescent inhalant drug involvement.

OBJECTIVE: To fill some of the gaps in our knowledge of the epidemiology of inhalant drug use. We examine age, sex, and race or ethnicity variations in the occurrence of inhalant use as well as time trends and the purported transitory nature of inhalant use among adolescents in the United States. DESIGN: The data analyzed in this report were collected as part of the National Household Survey on Drug Abuse from 1990 through 1995. This annual survey is designed to provide cross-sectional information about the patterns of drug use among nationally representative samples of US household residents aged 12 years and older. PARTICIPANTS: The sample included 34826 adolescents aged 12 to 17 years who participated in the National Household Survey on Drug Abuse from 1990 through 1995 (N=2177-8005 per survey year). RESULTS: The use of inhalants, particularly aerosols and glue, increased during the first half of the current decade. Initiation of inhalant use is not limited to early adolescence and is not a transitory behavior among adolescents in the United States. The use of inhalants is equally common among members of both sexes, and non-Hispanic white youths are more likely to report use than are members of other race or ethnic groups. CONCLUSIONS: The increasing use of inhalants, their widespread availability, and the risks involved with their use indicate a need for more focused attention on this public health problem.

Association of the ADH2*2 allele with reduced ethanol consumption in Jewish men in Israel: a pilot study.

OBJECTIVE: This study provides preliminary evidence on the associations between alcohol consumption patterns and polymorphisms of the alcohol dehydrogenase (ADH) and aldehyde dehydrogenase (ALDH) enzymes in a Jewish population. METHOD: Two groups of Jewish men were studied--one group (n = 92) representative of the free-living population of Jerusalem and generally light consumers of ethanol and the other group (n = 53) composed of treatment-enrolled heroin dependent individuals in the same city, most with a history of heavy daily drinking. All participants were interviewed regarding sociodemographic background, present and past alcohol consumption patterns, and familial characteristics including alcohol problems among first-degree relatives. Polymorphisms of the ADH2, ADH3 and ALDH loci were determined for all participants. RESULTS: The less common allele of the ADH2 locus (ADH2*2 allele frequency approximately 20% in Ashkenazic and non-Ashkenazic members of both groups) was related to a reduced mean level of peak weekly alcohol intake in the two groups. In multiple regression models adjusting for family history of alcohol problems and other factors, the ADH2*2 allele accounted for 20% and 30% of the explained alcohol intake variance in these two groups, respectively. Results from a logistic regression indicated that the ADH2*2 allele was also related to infrequent drinking in both groups. Evidence for an independent association between the ADH3 polymorphism and alcohol consumption patterns was not found. The ALDH gene was not polymorphic in this population. CONCLUSIONS: This report describes for the first time an association between alcohol consumption patterns and a polymorphism at the ADH2 locus in a Jewish population. The relatively high frequency of the ADH2*2 allele may contribute to the seemingly lower levels of alcohol consumption and heightened sensitivity to alcohol observed among Jews.

Initial opportunity to use marijuana and the transition to first use: United States, 1979-1994.

A renewed American interest in marijuana has coincided with our research group's focus on the earliest stages of drug involvement. Here, we have studied the transition from an initial opportunity to try marijuana to the subsequent use of this drug. We analyzed self-report interview data gathered from nationally representative samples of the United States National Household Surveys on Drug Abuse, 1979-1994. The evidence indicates that the estimated prevalence of an opportunity to try marijuana has been rather stable for 15 years. However, there are recent increases in the probability of rapidly progressing from first marijuana opportunity to first marijuana use, among persons given an opportunity to use. In addition, the transition from first marijuana opportunity to eventual marijuana use seems to depend upon age at first opportunity. This epidemiological evidence on the transition from marijuana opportunity to marijuana use, the first to be published based on a nationally representative US sample, highlights directions for future research and a focus for prevention efforts.

Childhood misbehavior and the risk of injecting drug use.

This investigation examined the association between misbehavior in early life and subsequent injecting drug use (IDU) among Baltimore participants in the NIMH Epidemiologic Catchment Area study. Information about early misbehavior was collected in 1981 and about IDU in 1994-1996. Conditional logistic regression models were used to estimate the degree of association between early misbehavior and IDU, with injecting drug users (n = 38) and controls matched on census tract of residence (n = 745), selected from within the same community survey sample, and assessed in a methodologically identical manner. Early misbehavior was associated with subsequent IDU, independent of the associations with sex, age, and race. The estimated risk of IDU increased with the number of reported conduct problems. Early misbehavior also discriminated between injecting drug users and non-injecting drug users. These results, which support and extend earlier findings, have public health implications for preventing and stopping the risk-laden practice of injecting drug use.